Skip to content

Commit

Permalink
Merge pull request #4316 from OpenMathLib/develop
Browse files Browse the repository at this point in the history
Merge develop into release-0.3.0 for 0.3.25
  • Loading branch information
martin-frbg committed Nov 12, 2023
2 parents 0e54cbd + 64c9671 commit 5e1a429
Show file tree
Hide file tree
Showing 491 changed files with 14,379 additions and 39,851 deletions.
10 changes: 10 additions & 0 deletions .cirrus.yml
Original file line number Diff line number Diff line change
Expand Up @@ -148,6 +148,16 @@ FreeBSD_task:
- ls -l /usr/local/lib
- gmake CC=gcc INTERFACE64=1

FreeBSD_task:
name: FreeBSD-clang-openmp
freebsd_instance:
image_family: freebsd-13-2
install_script:
- pkg update -f && pkg upgrade -y && pkg install -y gmake gcc
- ln -s /usr/local/lib/gcc12/libgfortran.so.5.0.0 /usr/lib/libgfortran.so
compile_script:
- gmake CC=clang FC=gfortran USE_OPENMP=1 CPP_THREAD_SAFETY_TEST=1

#task:
# name: Windows/LLVM16 --- too slow ---
# windows_container:
Expand Down
16 changes: 16 additions & 0 deletions .cirun.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
# Self-Hosted Github Action Runners on AWS via Cirun.io
# Reference: https://docs.cirun.io/reference/yaml
runners:
- name: "aws-runner-graviton"
# Cloud Provider: AWS
cloud: "aws"
region: "us-east-1"
# Cheapest VM on AWS
instance_type: "c7g.large"
# Ubuntu-22.04, ami image
machine_image: "ami-0a0c8eebcdd6dcbd0"
preemptible: false
# Add this label in the "runs-on" param in .github/workflows/<workflow-name>.yml
# So that this runner is created for running the workflow
labels:
- "cirun-aws-runner-graviton"
139 changes: 139 additions & 0 deletions .github/workflows/arm64_graviton.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,139 @@
name: arm64 graviton cirun

on:
push:
branches:
- develop
- release-**
pull_request:
branches:
- develop
- release-**

concurrency:
group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
cancel-in-progress: true

permissions:
contents: read # to fetch code (actions/checkout)

jobs:
build:
if: "github.repository == 'OpenMathLib/OpenBLAS'"
runs-on: "cirun-aws-runner-graviton--${{ github.run_id }}"

strategy:
fail-fast: false
matrix:
fortran: [gfortran]
build: [cmake, make]

steps:
- name: Checkout repository
uses: actions/checkout@v3

- name: Print system information
run: |
if [ "$RUNNER_OS" == "Linux" ]; then
cat /proc/cpuinfo
else
echo "::error::$RUNNER_OS not supported"
exit 1
fi
- name: Install Dependencies
run: |
if [ "$RUNNER_OS" == "Linux" ]; then
sudo apt update
sudo apt-get install -y gfortran cmake ccache libtinfo5
else
echo "::error::$RUNNER_OS not supported"
exit 1
fi
- name: Compilation cache
uses: actions/cache@v3
with:
path: ~/.ccache
# We include the commit sha in the cache key, as new cache entries are
# only created if there is no existing entry for the key yet.
# GNU make and cmake call the compilers differently. It looks like
# that causes the cache to mismatch. Keep the ccache for both build
# tools separate to avoid polluting each other.
key: ccache-${{ runner.os }}-${{ matrix.build }}-${{ matrix.fortran }}-${{ github.ref }}-${{ github.sha }}
# Restore a matching ccache cache entry. Prefer same branch and same Fortran compiler.
restore-keys: |
ccache-${{ runner.os }}-${{ matrix.build }}-${{ matrix.fortran }}-${{ github.ref }}
ccache-${{ runner.os }}-${{ matrix.build }}-${{ matrix.fortran }}
ccache-${{ runner.os }}-${{ matrix.build }}
- name: Configure ccache
run: |
if [ "${{ matrix.build }}" = "make" ]; then
# Add ccache to path
if [ "$RUNNER_OS" = "Linux" ]; then
echo "/usr/lib/ccache" >> $GITHUB_PATH
else
echo "::error::$RUNNER_OS not supported"
exit 1
fi
fi
# Limit the maximum size and switch on compression to avoid exceeding the total disk or cache quota (5 GB).
test -d ~/.ccache || mkdir -p ~/.ccache
echo "max_size = 300M" > ~/.ccache/ccache.conf
echo "compression = true" >> ~/.ccache/ccache.conf
ccache -s
- name: Build OpenBLAS
run: |
case "${{ matrix.build }}" in
"make")
make -j$(nproc) DYNAMIC_ARCH=1 USE_OPENMP=0 FC="ccache ${{ matrix.fortran }}"
;;
"cmake")
mkdir build && cd build
cmake -DDYNAMIC_ARCH=1 \
-DNOFORTRAN=0 \
-DBUILD_WITHOUT_LAPACK=0 \
-DCMAKE_VERBOSE_MAKEFILE=ON \
-DCMAKE_BUILD_TYPE=Release \
-DCMAKE_Fortran_COMPILER=${{ matrix.fortran }} \
-DCMAKE_C_COMPILER_LAUNCHER=ccache \
-DCMAKE_Fortran_COMPILER_LAUNCHER=ccache \
..
cmake --build .
;;
*)
echo "::error::Configuration not supported"
exit 1
;;
esac
- name: Show ccache status
continue-on-error: true
run: ccache -s

- name: Run tests
timeout-minutes: 60
run: |
case "${{ matrix.build }}" in
"make")
MAKE_FLAGS='DYNAMIC_ARCH=1 USE_OPENMP=0'
echo "::group::Tests in 'test' directory"
make -C test $MAKE_FLAGS FC="ccache ${{ matrix.fortran }}"
echo "::endgroup::"
echo "::group::Tests in 'ctest' directory"
make -C ctest $MAKE_FLAGS FC="ccache ${{ matrix.fortran }}"
echo "::endgroup::"
echo "::group::Tests in 'utest' directory"
make -C utest $MAKE_FLAGS FC="ccache ${{ matrix.fortran }}"
echo "::endgroup::"
;;
"cmake")
cd build && ctest
;;
*)
echo "::error::Configuration not supported"
exit 1
;;
esac
5 changes: 5 additions & 0 deletions .github/workflows/c910v.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,16 @@ name: c910v qemu test

on: [push, pull_request]

concurrency:
group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
cancel-in-progress: true

permissions:
contents: read # to fetch code (actions/checkout)

jobs:
TEST:
if: "github.repository == 'OpenMathLib/OpenBLAS'"
runs-on: ubuntu-latest
env:
xuetie_toolchain: https://occ-oss-prod.oss-cn-hangzhou.aliyuncs.com/resource//1663142514282
Expand Down
21 changes: 14 additions & 7 deletions .github/workflows/dynamic_arch.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,16 @@ name: continuous build

on: [push, pull_request]

concurrency:
group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
cancel-in-progress: true

permissions:
contents: read # to fetch code (actions/checkout)

jobs:
build:
if: "github.repository == 'OpenMathLib/OpenBLAS'"
runs-on: ${{ matrix.os }}

strategy:
Expand Down Expand Up @@ -146,18 +151,19 @@ jobs:
msys2:
if: "github.repository == 'OpenMathLib/OpenBLAS'"
runs-on: windows-latest

strategy:
fail-fast: false
matrix:
msystem: [MINGW64, MINGW32, CLANG64, CLANG32]
msystem: [UCRT64, MINGW32, CLANG64, CLANG32]
idx: [int32, int64]
build-type: [Release]
include:
- msystem: MINGW64
- msystem: UCRT64
idx: int32
target-prefix: mingw-w64-x86_64
target-prefix: mingw-w64-ucrt-x86_64
fc-pkg: fc
- msystem: MINGW32
idx: int32
Expand All @@ -175,10 +181,10 @@ jobs:
target-prefix: mingw-w64-clang-i686
fc-pkg: cc
c-lapack-flags: -DC_LAPACK=ON
- msystem: MINGW64
- msystem: UCRT64
idx: int64
idx64-flags: -DBINARY=64 -DINTERFACE64=1
target-prefix: mingw-w64-x86_64
target-prefix: mingw-w64-ucrt-x86_64
fc-pkg: fc
- msystem: CLANG64
idx: int64
Expand All @@ -188,9 +194,9 @@ jobs:
# Compiling with Flang 16 seems to cause test errors on machines
# with AVX512 instructions. Revisit after MSYS2 distributes Flang 17.
no-avx512-flags: -DNO_AVX512=1
- msystem: MINGW64
- msystem: UCRT64
idx: int32
target-prefix: mingw-w64-x86_64
target-prefix: mingw-w64-ucrt-x86_64
fc-pkg: fc
build-type: None
exclude:
Expand Down Expand Up @@ -312,6 +318,7 @@ jobs:
cross_build:
if: "github.repository == 'OpenMathLib/OpenBLAS'"
runs-on: ubuntu-22.04

strategy:
Expand Down
8 changes: 8 additions & 0 deletions .github/workflows/loongarch64.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,8 +2,13 @@ name: loongarch64 qemu test

on: [push, pull_request]

concurrency:
group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
cancel-in-progress: true

jobs:
TEST:
if: "github.repository == 'OpenMathLib/OpenBLAS'"
runs-on: ubuntu-latest
strategy:
fail-fast: false
Expand All @@ -18,6 +23,9 @@ jobs:
- target: LOONGSON2K1000
triple: loongarch64-unknown-linux-gnu
opts: NO_SHARED=1 TARGET=LOONGSON2K1000
- target: DYNAMIC_ARCH
triple: loongarch64-unknown-linux-gnu
opts: NO_SHARED=1 DYNAMIC_ARCH=1 TARGET=GENERIC

steps:
- name: Checkout repository
Expand Down
5 changes: 5 additions & 0 deletions .github/workflows/mips64.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,16 @@ name: mips64 qemu test

on: [push, pull_request]

concurrency:
group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
cancel-in-progress: true

permissions:
contents: read # to fetch code (actions/checkout)

jobs:
TEST:
if: "github.repository == 'OpenMathLib/OpenBLAS'"
runs-on: ubuntu-latest
strategy:
fail-fast: false
Expand Down
5 changes: 5 additions & 0 deletions .github/workflows/nightly-Homebrew-build.yml
Original file line number Diff line number Diff line change
Expand Up @@ -18,11 +18,16 @@ on:

name: Nightly-Homebrew-Build

concurrency:
group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
cancel-in-progress: true

permissions:
contents: read # to fetch code (actions/checkout)

jobs:
build-OpenBLAS-with-Homebrew:
if: "github.repository == 'OpenMathLib/OpenBLAS'"
runs-on: macos-latest
env:
DEVELOPER_DIR: /Applications/Xcode_11.4.1.app/Contents/Developer
Expand Down
46 changes: 46 additions & 0 deletions Changelog.txt
Original file line number Diff line number Diff line change
@@ -1,4 +1,50 @@
OpenBLAS ChangeLog
====================================================================
Version 0.3.25
12-Nov-2023

general:
- improved the error message shown on exceeding the maximum thread count
- improved the code to add supplementary thread buffers in case of overflow
- fixed a potential division by zero in ?ROTG
- improved the ?MATCOPY functions to accept zero-sized rows or columns
- corrected empty prototypes in function declarations
- cleaned up unused declarations in the f2c-converted versions of the LAPACK sources
- fixed compilation with the Cray CCE Compiler suite
- improved link line rewriting to avoid mixed libgomp/libomp builds with clang&gfortran
- worked around OPENMP builds with LLVM14's libomp hanging on FreeBSD
- improved the Makefiles to require less option duplication on "make install"
- imported the following changes from the upcoming release 3.12 of Reference-LAPACK
- deprecate utility functions ?GELQS and ?GEQRS (LAPACK PR 900)
- apply rounding up to workspace calculations done in floating point (LAPACK PR 904)
- avoid overflow in STGEX2/DTGEX2 (LAPACK PR 907)
- fix accumulation in ?LASSQ (LAPACK PR 909)
- fix handling of NaN values in ?GECON (LAPACK PR 926)
- avoid overflow in CBDSQR/ZBDSQR (LAPACK PR 927)
- fix poor vector orthogonalizations in ?ORBDB5/?UNBDB5 (LAPACK PR 928 & 930)

x86-64:
- fixed compile-time autodetection of AMD Ryzen3 and Ryzen4 cpus
- fixed capability-based fallback selection for unknown cpus in DYNAMIC_ARCH
- added AVX512 optimizations for ?ASUM on Sapphire Rapids and Cooper Lake

ARM64:
- fixed building on Apple with homebrew gcc
- fixed building with XCODE 15
- fixed building on A64FX and Cortex A710/X1/X2
- increased the default buffer size for recent ARM server cpus

POWER:
- fixed building with the IBM xlf 16.1.1 compiler
- fixed building with IBM XL C
- added support for DYNAMIC_ARCH builds with clang
- fixed union declaration in the BFLOAT16 test case
- enable optimizations for the AIX assembler on POWER10

LOONGARCH64:
- added an optimized SGEMV kernel
- added an optimized DTRSM kernel

====================================================================
Version 0.3.24
03-Sep-2023
Expand Down

0 comments on commit 5e1a429

Please sign in to comment.